利用半學生氏t檢定偵測異質性疾病基因表現差異

Detecting Differentially Expressed Genes in Heterogeneous Disease Using Half Student t test

許郡倫(Chun-Lun Hsu)

指導教授 ： 李文宗

國立臺灣大學/公共衛生學院/流行病學研究所/碩士(2009年)

https://doi.org/10.6342/NTU.2009.00191

摘要

基因表現研究為近年來熱門的研究題材。研究者常利用學生氏t檢定篩選出與疾病相關的基因。然而，當研究者所研究的疾病為異質性疾病時，病例組和對照組間基因表現平均值的差距可能不大，因而較難被傳統的學生氏t檢定偵測出來。本研究提出半學生氏t檢定(half Student t test)處理異質性疾病的情況。半學生氏t檢定的檢定統計量只考慮對照組的樣本標準差，而不考慮病例組的樣本標準差。作者以蒙地卡羅模擬及一個結腸腫瘤基因表現真實資料，比較半學生氏t檢定與傳統學生氏t檢定的統計檢力表現。在本研究模擬的情境下，半學生氏t檢定最多可比傳統學生氏t檢定多出約35%的統計檢力。另外，結腸腫瘤基因表現資料經過錯誤發現率(切點訂為0.05)的控制，半學生氏t檢定可比傳統學生氏t檢定多偵測出279個顯著基因。本研究所提出的半學生氏t檢定執行容易且統計檢力表現良好，值得推荐做為針對異質性疾病偵測基因表現差異的方法。

關鍵字

學生氏t檢定 ； 基因表現 ； 異質性疾病 ； 檢力

並列摘要

Gene-expression has been a popular research topic in recent years. Student t-test is commonly adopted to screen disease-related genes. However, when the researches are focused on heterogeneous disease, the means of gene-expression levels between case group and control group may be similar, and thus, the difference would be difficult to be detected by conventional Student t-test. This study proposed half Student t-test to examine heterogeneous disease. Test statistics of half Student t-test only considers sample standard deviation of control group, without considering the sample standard deviation of case group. This study applied Monte Carlo simulation and real gene-expression data of colon cancer to compare the power performance of half Student t-test and conventional Student t-test. Under the simulated scenario, this study found that half Student t-test could have 35% higher statistical power than conventional Student t-test. In addition, after false discovery rate (cut-off point set at 0.05) control of colon cancer gene-expression data, half Student t-test could detect 279 more significant genes than conventional Student t-test. Half Student t-test is easy to execute with good statistical power, and is worth to be recommended as a method of detecting heterogeneous disease gene-expression difference.

並列關鍵字

Student t-test ； gene-expression ； heterogeneous disease ； power

參考文獻

1. Allison DB, Cui X, Page GP, et al. Microarray data analysis: from disarray to consolidation and consensus. Nat Rev Genet 2006;7:55-65.

2. Sandrine Dudoit YHY, Matthew J. Callow, Terence P. Speed Statistical methods for identifying differentially expressed genes in replicated cDNA microarray experiments. Statistica Sinica 2002;12:111-39.

3. Pan W. A comparative review of statistical methods for discovering differentially expressed genes in replicated microarray experiments. Bioinformatics 2002;18:546-54.

4. van der Pouw Kraan TC, van Gaalen FA, Kasperkovitz PV, et al. Rheumatoid arthritis is a heterogeneous disease: evidence for differences in the activation of the STAT-1 pathway between rheumatoid tissues. Arthritis Rheum 2003;48:2132-45.

5. Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000;403:503-11.